Liver FACS Notebook

Enter the name of the tissue you want to analyze.

tissue_of_interest = "Liver"

Install any packages you’re missing.

#install.packages("useful")
#install.packages("ontologyIndex")
#install.packages("here")

Load the requisite packages and some additional helper functions.

library(here)
library(useful)
library(Seurat)
library(dplyr)
library(Matrix)
library(ontologyIndex)
cell_ontology = get_ontology('https://raw.githubusercontent.com/obophenotype/cell-ontology/master/cl-basic.obo', extract_tags='everything')
validate_cell_ontology = function(cell_ontology_class){
  in_cell_ontology = sapply(cell_ontology_class, function(x) is.element(x, cell_ontology$name) || is.na(x))
  if (!all(in_cell_ontology)) {
    message = paste0('"', cell_ontology_class[!in_cell_ontology], '" is not in the cell ontology\n')
    stop(message)
  }
}
convert_to_cell_ontology_id = function(cell_ontology_class){
  return(sapply(cell_ontology_class, function(x) {
      if(is.na(x)){
        x
      }else{
        as.vector(cell_ontology$id[cell_ontology$name == x])[1]
      }
    }))
}

# Load the per-plate metadata
plate_metadata_filename = here('00_data_ingest', '00_facs_raw_data', 'metadata_FACS.csv')
plate_metadata <- read.csv(plate_metadata_filename, sep=",", header = TRUE)
colnames(plate_metadata)[1] <- "plate.barcode"
plate_metadata

Load the read count data.

#Load the gene names and set the metadata columns by opening the first file
filename = here('00_data_ingest', '00_facs_raw_data', 'FACS', paste0(tissue_of_interest, '-counts.csv'))
raw.data = read.csv(filename, sep=",", row.names=1)
# raw.data = data.frame(row.names = rownames(raw.data))
corner(raw.data)

Make a vector of plate barcodes for each cell

plate.barcodes = lapply(colnames(raw.data), function(x) strsplit(strsplit(x, "_")[[1]][1], '.', fixed=TRUE)[[1]][2])

Make per-cell metadata, and reorder the raw data by plate for consistency. Make a plate barcode dataframe to “expand” the per-plate metadata to be per-cell.

barcode.df = t.data.frame(as.data.frame(plate.barcodes))
rownames(barcode.df) = colnames(raw.data)
colnames(barcode.df) = c('plate.barcode')
head(barcode.df)

                        plate.barcode
F18.MAA000377.3_9_M.1.1 "MAA000377"  
J20.MAA000377.3_9_M.1.1 "MAA000377"  
F19.MAA000377.3_9_M.1.1 "MAA000377"  
J21.MAA000377.3_9_M.1.1 "MAA000377"  
F20.MAA000377.3_9_M.1.1 "MAA000377"  
J22.MAA000377.3_9_M.1.1 "MAA000377"  

rnames = row.names(barcode.df)
head(rnames)

[1] "F18.MAA000377.3_9_M.1.1" "J20.MAA000377.3_9_M.1.1" "F19.MAA000377.3_9_M.1.1" "J21.MAA000377.3_9_M.1.1" "F20.MAA000377.3_9_M.1.1"
[6] "J22.MAA000377.3_9_M.1.1"

meta.data <- merge(barcode.df, plate_metadata, by='plate.barcode', sort = F)
row.names(meta.data) <- rnames
head(meta.data)

# Sort cells by plate barcode because that's how the data was originally
meta.data = meta.data[order(rownames(meta.data)), ]
corner(meta.data)

raw.data = raw.data[, rownames(meta.data)]
corner(raw.data)

Process the raw data and load it into the Seurat object.

# Find ERCC's, compute the percent ERCC, and drop them from the raw data.
erccs <- grep(pattern = "^ERCC-", x = rownames(x = raw.data), value = TRUE)
percent.ercc <- Matrix::colSums(raw.data[erccs, ])/Matrix::colSums(raw.data)
ercc.index <- grep(pattern = "^ERCC-", x = rownames(x = raw.data), value = FALSE)
raw.data <- raw.data[-ercc.index,]
# Create the Seurat object with all the data
tiss <- CreateSeuratObject(raw.data = raw.data, project = tissue_of_interest, 
                    min.cells = 1, min.genes = 0)
tiss <- AddMetaData(object = tiss, meta.data)
tiss <- AddMetaData(object = tiss, percent.ercc, col.name = "percent.ercc")
# Change default name for sums of counts from nUMI to nReads
colnames(tiss@meta.data)[colnames(tiss@meta.data) == 'nUMI'] <- 'nReads'
# Create metadata columns for cell_ontology_classs and subcell_ontology_classs
tiss@meta.data[,'free_annotation'] <- NA
tiss@meta.data[,'free_annotation'] <- NA
tiss@meta.data[,'cell_ontology_class'] <- NA

Calculate percent ribosomal genes.

ribo.genes <- grep(pattern = "^Rp[sl][[:digit:]]", x = rownames(x = tiss@data), value = TRUE)
percent.ribo <- Matrix::colSums(tiss@raw.data[ribo.genes, ])/Matrix::colSums(tiss@raw.data)
tiss <- AddMetaData(object = tiss, metadata = percent.ribo, col.name = "percent.ribo")

A sanity check: genes per cell vs reads per cell.

GenePlot(object = tiss, gene1 = "nReads", gene2 = "nGene", use.raw=T)

Filter out cells with few reads and few genes.

tiss <- FilterCells(object = tiss, subset.names = c("nGene", "nReads"), low.thresholds = c(500, 50000))

Normalize the data, then center and scale.

tiss <- NormalizeData(object = tiss, scale.factor = 1e6)

Performing log-normalization
0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
**************************************************|

tiss <- ScaleData(object = tiss)

[1] "Scaling data matrix"

  |                                                                                                                                           
  |                                                                                                                                     |   0%
  |                                                                                                                                           
  |=====================================================================================================================================| 100%

tiss <- FindVariableGenes(object = tiss, do.plot = TRUE, x.high.cutoff = Inf, y.cutoff = 0.5)

Calculating gene means
0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
**************************************************|
Calculating gene variance to mean ratios
0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
**************************************************|

Run Principal Component Analysis.

tiss <- RunPCA(object = tiss, do.print = FALSE)
tiss <- ProjectPCA(object = tiss, do.print = FALSE)

Later on (in FindClusters and TSNE) you will pick a number of principal components to use. This has the effect of keeping the major directions of variation in the data and, ideally, supressing noise. There is no correct answer to the number to use, but a decent rule of thumb is to go until the plot plateaus.

PCElbowPlot(object = tiss)

Choose the number of principal components to use.

# Set number of principal components. 
n.pcs = 11

The clustering is performed based on a nearest neighbors graph. Cells that have similar expression will be joined together. The Louvain algorithm looks for groups of cells with high modularity–more connections within the group than between groups. The resolution parameter determines the scale. Higher resolution will give more clusters, lower resolution will give fewer.

For the top-level clustering, aim to under-cluster instead of over-cluster. It will be easy to subset groups and further analyze them below.

# Set resolution 
res.used <- 1
tiss <- FindClusters(object = tiss, reduction.type = "pca", dims.use = 1:n.pcs, 
    resolution = res.used, print.output = 0, save.SNN = TRUE)

We use TSNE solely to visualize the data.

# If cells are too spread out, you can raise the perplexity. If you have few cells, try a lower perplexity (but never less than 10).
tiss <- RunTSNE(object = tiss, dims.use = 1:n.pcs, seed.use = 10, perplexity=30)

TSNEPlot(object = tiss, do.label = T, pt.size = 1.2, label.size = 4)

Check expression of genes useful for indicating cell type. For the islet cells, the mRNA for their specific secretory molecule is a strong signal.

hepatocyte: Alb, Ttr, Apoa1, and Serpina1c pericentral: Cyp2e1, Glul, Oat, Gulo midlobular: Ass1, Hamp, Gstp1, Ubb periportal: Cyp2f2, Pck1, Hal, Cdh1

endothelial cells: Pecam1, Nrp1, Kdr+ and Oit3+ Kuppfer cells: Emr1, Clec4f, Cd68, Irf7 NK/NKT cells: Zap70, Il2rb, Nkg7, Cxcr6, Klr1c, Gzma B cells: Cd79a, Cd79b, Cd74 and Cd19

genes_hep = c('Alb', 'Ttr', 'Apoa1', 'Serpina1c', 
                   'Cyp2e1', 'Glul', 'Oat', 'Gulo',
                   'Ass1', 'Hamp', 'Gstp1', 'Ubb',
                   'Cyp2f2', 'Pck1', 'Hal', 'Cdh1')
genes_endo = c('Pecam1', 'Nrp1', 'Kdr','Oit3')
genes_kuppfer = c('Emr1', 'Clec4f', 'Cd68', 'Irf7')
genes_nk = c('Zap70', 'Il2rb', 'Nkg7', 'Cxcr6', 'Gzma')
genes_b = c('Cd79a', 'Cd79b', 'Cd74')
all_genes = c(genes_hep, genes_endo, genes_kuppfer, genes_nk, genes_b)

Dotplots let you see the intensity of exppression and the fraction of cells expressing for each of your genes of interest. The radius shows you the percent of cells in that cluster with at least one read sequenced from that gene. The color level indicates the average Z-score of gene expression for cells in that cluster, where the scaling is done over taken over all cells in the sample.

Violin plot of Albumin expression to assess possible leakage.

We can also find all differentially expressed genes marking each cluster. This may take some time.

#clust.markers0 <- FindMarkers(object = tiss, ident.1 = 0, only.pos = TRUE, min.pct = 0.25, thresh.use = 0.25)
#tiss.markers <- FindAllMarkers(object = tiss, only.pos = TRUE, min.pct = 0.25, thresh.use = 0.25)

Display the top markers you computed above.

#tiss.markers %>% group_by(cluster) %>% top_n(5, avg_diff)

Using the markers above, we can confidentaly label many of the clusters:

tiss <- StashIdent(object = tiss, save.name = "cluster.ids")
cluster.ids <- c(0, 1, 2, 3, 4, 5, 6, 7, 8)
free_annotation <- c(
  "endothelial cell", 
  "hepatocyte", 
  "hepatocyte",
  "hepatocyte",
  "hepatocyte",
  "kuppfer",
  "hepatocyte",
  "B cell",
  "Nk/NKT cells")
cell_ontology_class <-c(
  "endothelial cell of hepatic sinusoid", 
  "hepatocyte", 
  "hepatocyte",
  "hepatocyte",
  "hepatocyte",
  "Kupffer cell",
  "hepatocyte",
  "B cell",
  "natural killer cell")
validate_cell_ontology(cell_ontology_class)
cell_ontology_id = convert_to_cell_ontology_id(cell_ontology_class)
tiss@meta.data['free_annotation'] <- as.character(plyr::mapvalues(x = tiss@ident, from = cluster.ids, to = free_annotation))
validate_cell_ontology(cell_ontology_class)
cell_ontology_id = convert_to_cell_ontology_id(cell_ontology_class)
tiss@meta.data['cell_ontology_class'] <- as.character(plyr::mapvalues(x = tiss@ident, from = cluster.ids, to = cell_ontology_class))
tiss@meta.data['cell_ontology_id'] <- as.character(plyr::mapvalues(x = tiss@ident, from = cluster.ids, to = cell_ontology_id))

Checking for batch effects

Color by metadata, like plate barcode, to check for batch effects.

TSNEPlot(object = tiss, do.return = TRUE, group.by = "mouse.id")

table(FetchData(tiss, c('plate.barcode','ident')) %>% droplevels())

             ident
plate.barcode   0   1   2   3   4   5   6   7   8
    MAA000377  22   0  22   5   0  14   0  10   7
    MAA000907 160   0  36   2   0  47   0  31  32
    MAA100039   0   0  13  21  31   0   0   0   0
    MAA100040   0   0  10   6  20   0   0   0   0
    MAA100041   0   0   3   9  10   0   0   0   0
    MAA100042   0   0   7  23  11   0   1   0   0
    MAA100138   0  46   0  11   0   0  25   0   0
    MAA100140   0  46   0   5   0   0  28   0   0

Final coloring

Color by cell ontology class on the original TSNE.

TSNEPlot(object = tiss, group.by = "cell_ontology_class")

Save the Robject for later

filename = here('00_data_ingest', '04_tissue_robj_generated', 
                     paste0("facs_", tissue_of_interest, "_seurat_tiss.Robj"))
print(filename)

[1] "/Users/josh/src/tabula-muris/00_data_ingest/04_tissue_robj_generated/facs_Liver_seurat_tiss.Robj"

save(tiss, file=filename)

# To reload a saved object
#filename = here('00_data_ingest', '04_tissue_robj_generated', 
#                      paste0("facs_", tissue_of_interest, "_seurat_subtiss.Robj"))
#load(file=filename)

BuildClusterTree(tiss) 

[1] "Finished averaging RNA for cluster 0"
[1] "Finished averaging RNA for cluster 1"
[1] "Finished averaging RNA for cluster 2"
[1] "Finished averaging RNA for cluster 3"
[1] "Finished averaging RNA for cluster 4"
[1] "Finished averaging RNA for cluster 5"
[1] "Finished averaging RNA for cluster 6"
[1] "Finished averaging RNA for cluster 7"
[1] "Finished averaging RNA for cluster 8"
An object of class seurat in project Liver 
 17448 genes across 714 samples.

Export the final metadata

So that Biohub can easily combine all your cell_ontology_classs, please export them as a simple csv.

filename = here('00_data_ingest', '03_tissue_annotation_csv', 
                     paste0(tissue_of_interest, "_annotation.csv"))
write.csv(FetchData(tiss, c('plate.barcode','cell_ontology_class','cell_ontology_id', 'free_annotation', 'tSNE_1', 'tSNE_2')), file=filename)

Figures for Supplement